• No results found

Research data management and the Flemish Universities

N/A
N/A
Protected

Academic year: 2022

Share "Research data management and the Flemish Universities"

Copied!
14
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

V L I R Vlaamse

Interuniversitaire Raad

Research data management and the Flemish Universities

VLIR WG Research Data Management & Open Science

(2)

V L I R Vlaamse

Interuniversitaire Raad

and ultimately Open Science

Research data management, an essential step towards Open Data

FLEMISH UNIVERSITIES WORKING TOGETHER

(3)

V L I R Vlaamse

Interuniversitaire Raad

Collaboration on practical issues:

DMPonline.be DMP templates RDM Survey:

State of affairs in Flanders, depict RDM needs &

requirements

Onderzoeksmanagement aan de Vlaamse Universiteiten,

TH&MA (3), 47-51, 2017

Joint Policy statements:

Research data management en de Vlaamse Universiteiten:

White paper, VLIR-WG RDM &

Open Science

Research data management, an essential step towards Open Data

FLEMISH UNIVERSITIES WORKING TOGETHER

(4)

V L I R Vlaamse

Interuniversitaire Raad

Questions

Consistent with foreign surveys so that

benchmarking is possible

Two tier system:

Survey to principal investigators (n=224)

Survey to researchers (n=425)

RDM @ the Flemish universities: State of affairs

5 dimensions:

• Data formats

• Archiving, storage

• Ethical & Legal

• Infrastructure & Services

• Collaboration and Reuse

Survey set-up

(5)

V L I R Vlaamse

Interuniversitaire Raad

Policies

AD HOC RATHER THAN ORGANISED

34%

26%

21%

11%

8%

DO YOU MAKE A DATA MANAGEMENT PLAN BEFORE THE START OF A

PROJECT?

Never Hardly ever Mostly Yes, always Yes, when requested

5%

38%

57%

DATA MANAGEMENT POLICY

Formal Informal Non

(6)

V L I R Vlaamse

Interuniversitaire Raad

Access

MORE BASED ON “NEED TO” THAN ON “WANT TO”

0 20 40 60 80 100 120 140

Other The public The scientific community No one All members of my institution Peer reviewers Interested persons by request Selected members of my institution

DATA ACCESSIBILITY: TO WHOM

0 20 40 60 80 100 120 140

Not at all Other Via personal or institutional website As linked supplementary material for publications Via data archive/repository Via remote server or share drives Via physical disks Via cloud applications Via email

DATA ACCESSIBILITY: MEDIUM

(7)

V L I R Vlaamse

Interuniversitaire Raad

What is your opinion on storing (a selection of)

the research data of your unit in an open access data repository

so that they can be reused by other

researchers?

Opinions concerning openness

DIVIDED BUT HOPEFUL

49% 51%

OPEN ACCESS DATA

Preferred Not preferred

(8)

V L I R Vlaamse

Interuniversitaire Raad

Sharing

ADVANTAGES ARE RECOGNISED BUT LEGAL ISSUES AND LACK OF TIME

0 10 20 30 40 50 60 70 80

Use of rare data formats Other Missing data processes Missing data standards Lack of motivation for sharing Potentially undesired commercial use Lack of infrastructure Risk of misinterpretation and/or falsification of data Danger of misuse Other legal restrictions (e.g. copyright, patent law, …

Increased competition in the „publish or perish“ … Increased effort of time and/or cost

Privacy violation

WHAT KEEPS YOU FROM SHARING DATA?

0 20 40 60 80 100 120 140

Other None Financial incentives (bonus, expense allowance) Support in the process of making the data accessible Establishment of standards for accountability and …

User friendly infrastructure Recognition in the scientific community Consideration of research data as relevant scientific …

New contacts and/or opportunities for cooperation … Increased visibility and impact of your own research …

OPEN DATA SHARING INCENTIVES

(9)

V L I R Vlaamse

Interuniversitaire Raad

And ultimately Open Science

Research data management, an essential step towards Open Data

FLEMISH UNIVERSITIES WORKING TOGETHER

(10)

V L I R Vlaamse

Interuniversitaire Raad

And ultimately Open Science

Research data management, an essential step towards Open Data

FLEMISH UNIVERSITIES WORKING TOGETHER

Massive change of the way research works

Is it Utopia or can RDM be facilitated?

(11)

V L I R Vlaamse

Interuniversitaire Raad

Research Data Management

WHAT IF NOTHING IS BEING DONE?

1. In 2020 the market share of open data in EU28+ will have risen to 75.7 billion euro and an estimated 100.000 data steward positions need to be filled

2. By 2020 an estimated 1.7 billion € net gain through increased efficiency in information provision

i.e. No redundant analyses, rapid sharing  saves money, time and sometimes even lives

3. Opening and sharing data fosters more visibility & impact in science and beyond

4. Are commercial providers the best partners in ensuring fair & sustainable access to our data Alternative 1: Are we prepared to lose our competitive edge, money, time, trust & governance?

Alternative 2: Will Flanders take this opportunity to boost knowledge and innovation in the region?

(12)

V L I R Vlaamse

Interuniversitaire Raad

Research Data Management

RECOMMENDATIONS

1. INVEST IN INFRASTRUCTURE

1. Data = intellectual capital

2. EOSC is great, if your data repository is interoperable with this service 3. Data services: 1.000.000 £ set up, 500.000 £ recurrent cost

Q: What data repositories?

Q: Joint supported initiative (DANS) or multiply costs for all research institutions (i.e. 5X Flem. universities,…)

2. INVEST IN EDUCATION

1. RDM skills are lacking: 100.000 data stewards, researchers, …

Q: If nobody has the skils, how can we make RDM work in a qualitative manner?

(13)

V L I R Vlaamse

Interuniversitaire Raad

Research Data Management

RECOMMENDATIONS

3. PROVIDE CLEAR & UNIFORM LEGISLATION

1. GDPR  Belgian law

2. EU Copyright & TDM reforms

Q: Don’t make it complicated for researchers to do research Q: Provide support

4. PROVIDE INCENTIVES FOR OPEN SCIENCE

1. There’s more to (academic) life then publications, IF-factors,… Data = intellectual capital 2. Open data is better than sharing data, sharing data is better than dead data

Q: Provide incentives, re-think evaluation models for research(ers)

(14)

V L I R Vlaamse

Interuniversitaire Raad

9-11-2017 | 14

danielle.gilliot@vlir.be nele.robberechts@vlir.be The collaborating members

With the kind support of

KU Leuven (Hannelore Vanhaverbeke; Joke Claeys) UAntwerpen (Jord Hanus; Marianne De Voecht)

UGent (Inge Van Nieuwerburgh; Myriam Mertens)

UHasselt (Sadia Vancauwenbergh; Hanne Elsen)

VUB (Lucy Amez; Kyle Van Gaeveren)

Referenties

GERELATEERDE DOCUMENTEN

Maar omdat de robot bij iedere melking tijd nodig heeft voor onder meer koeherkenning, speenreiniging en aansluiten van de tepelbekers, gaat een korter interval

Inspired by Ka- makura & Wedel (2000), a general framework based on latent variable models is proposed to analyze missing data. With this framework, the authors develop

In 2015, the Research Institute for Nature and Forest (INBO) adopted an open data policy, with the goal to publish our biodiversity data as open data, so anyone can use these.. In

similar to Section 6.2.3, conditional regression models can be specified for the audit values which do not coincide with the book value or previous audit values according to

Item scores are missing completely at random (MCAR; see Little & Rubin, 1987, pp. 14-17) if the cause of missingness is unrelated to the missing values themselves, the scores on

[r]

A broad variety of orthogonally reactive functionalities for cyclic monomers for the ROP and post-modification opportu- nities has been reported so far, which give access to

Invasive breast cancer The hospital organizational factors hospital type, hospital volume, percentage of mastectomies, number of weekly MDT meetings, number of plastic surgeons per